Optimal Design of an Axial Flux Permanent Magnet Synchronous Motor for the Electric Bicycle

Abstract

Design of an electric machine such as the axial flux permanent magnet synchronous motor (AFPMSM) requires a 3-D finite-element method (FEM) analysis. The AFPMSM with a 3-D FEM model involves too much time and effort to analyze. To deal with this problem, we apply a surrogate assisted multi-objective optimization (SAMOO) algorithm that can realize an accurate and well-distributed Pareto front set with a few number of function calls, and considers various design variables in the motor design process. The superior performance of the SAMOO is verified by comparing it with conventional multi-objective optimization algorithms in a test function. Finally, the optimal design result of the AFPMSM for the electric bicycle is obtained by using the SAMOO algorithm.